Combined PD-1, BRAF and MEK inhibition in BRAFV600E colorectal cancer: a phase 2 trial.

While BRAF inhibitor combinations with EGFR and/or MEK inhibitors have improved clinical efficacy in BRAFV600E colorectal cancer (CRC), response rates remain low and lack durability. Preclinical data suggest that BRAF/MAPK pathway inhibition may augment the tumor immune response. We performed a proof-of-concept single-arm phase 2 clinical trial of combined PD-1, BRAF and MEK inhibition with sparatlizumab (PDR001), dabrafenib and trametinib in 37 patients with BRAFV600E CRC. The primary end point was overall response rate, and the secondary end points were progression-free survival, disease control rate, duration of response and overall survival. The study met its primary end point with a confirmed response rate (24.3% in all patients; 25% in microsatellite stable patients) and durability that were favorable relative to historical controls of BRAF-targeted combinations alone. Single-cell RNA sequencing of 23 paired pretreatment and day 15 on-treatment tumor biopsies revealed greater induction of tumor cell-intrinsic immune programs and more complete MAPK inhibition in patients with better clinical outcome. Immune program induction in matched patient-derived organoids correlated with the degree of MAPK inhibition. These data suggest a potential tumor cell-intrinsic mechanism of cooperativity between MAPK inhibition and immune response, warranting further clinical evaluation of optimized targeted and immune combinations in CRC. ClinicalTrials.gov registration: NCT03668431.

Generalizing Few-Shot Classification of Whole-Genome Doubling Across Cancer Types.

The study and treatment of cancer is traditionally specialized to the cancer's site of origin. However, certain phenotypes are shared across cancer types and have important implications for clinical care. To date, automating the identification of these characteristics from routine clinical data - irrespective of the type of cancer - is impaired by tissue-specific variability and limited labeled data. Whole-genome doubling is one such phenotype; whole-genome doubling events occur in nearly every type of cancer and have significant prognostic implications. Using digitized histopathology slide images of primary tumor biopsies, we train a deep neural network end-to-end to accurately generalize few-shot classification of whole-genome doubling across 17 cancer types. By taking a meta-learning approach, cancer types are treated as separate but jointly-learned tasks. This approach outperforms a traditional neural network classifier and quickly generalizes to both held-out cancer types and batch effects. These results demonstrate the unrealized potential for meta-learning to not only account for between-cancer type variability but also remedy technical variability, enabling real-time identification of cancer phenotypes that are too often costly and inefficient to obtain.

Identifying Cell Type-Specific Chemokine Correlates with Hierarchical Signal Extraction from Single-Cell Transcriptomes.

Biological data is inherently heterogeneous and high-dimensional. Single-cell sequencing of transcripts in a tissue sample generates data for thousands of cells, each of which is characterized by upwards of tens of thousands of genes. How to identify the subsets of cells and genes that are associated with a label of interest remains an open question. In this paper, we integrate a signal-extractive neural network architecture with axiomatic feature attribution to classify tissue samples based on single-cell gene expression profiles. This approach is not only interpretable but also robust to noise, requiring just 5% of genes and 23% of cells in an in silico tissue sample to encode signal in order to distinguish signal from noise with greater than 70% accuracy. We demonstrate its applicability in two real-world settings for discovering cell type-specific chemokine correlates: predicting response to immune checkpoint inhibitors in multiple tissue types and classifying DNA mismatch repair status in colorectal cancer. Our approach not only significantly outperforms traditional machine learning classifiers but also presents actionable biological hypotheses of chemokinemediated tumor immunogenicity.

Spatially organized multicellular immune hubs in human colorectal cancer.

Immune responses to cancer are highly variable, with mismatch repair-deficient (MMRd) tumors exhibiting more anti-tumor immunity than mismatch repair-proficient (MMRp) tumors. To understand the rules governing these varied responses, we transcriptionally profiled 371,223 cells from colorectal tumors and adjacent normal tissues of 28 MMRp and 34 MMRd individuals. Analysis of 88 cell subsets and their 204 associated gene expression programs revealed extensive transcriptional and spatial remodeling across tumors. To discover hubs of interacting malignant and immune cells, we identified expression programs in different cell types that co-varied across tumors from affected individuals and used spatial profiling to localize coordinated programs. We discovered a myeloid cell-attracting hub at the tumor-luminal interface associated with tissue damage and an MMRd-enriched immune hub within the tumor, with activated T cells together with malignant and myeloid cells expressing T cell-attracting chemokines. By identifying interacting cellular programs, we reveal the logic underlying spatially organized immune-malignant cell networks.

A dominant-negative effect drives selection of TP53 missense mutations in myeloid malignancies.

TP53, which encodes the tumor suppressor p53, is the most frequently mutated gene in human cancer. The selective pressures shaping its mutational spectrum, dominated by missense mutations, are enigmatic, and neomorphic gain-of-function (GOF) activities have been implicated. We used CRISPR-Cas9 to generate isogenic human leukemia cell lines of the most common TP53 missense mutations. Functional, DNA-binding, and transcriptional analyses revealed loss of function but no GOF effects. Comprehensive mutational scanning of p53 single-amino acid variants demonstrated that missense variants in the DNA-binding domain exert a dominant-negative effect (DNE). In mice, the DNE of p53 missense variants confers a selective advantage to hematopoietic cells on DNA damage. Analysis of clinical outcomes in patients with acute myeloid leukemia showed no evidence of GOF for TP53 missense mutations. Thus, a DNE is the primary unit of selection for TP53 missense mutations in myeloid malignancies.